AI Adoption in Financial Crime & Compliance: 6 Key Takeaways

AI Adoption in Financial Crime & Compliance: 6 Key Takeaways

Hawk partnered with Chartis to survey 250 banks, payment firms, and fintechs to get clarity on the state of AI adoption and maturity. In this article, our Senior Product Marketing Manager, Erica Brackman, shares her three key findings from each report and what financial crime and compliance (FCC) leaders should take away from these findings. Read the banking report here and payments/fintech report here. 

 

What FCC leaders at banks need to know

 

AI is now table stakes in bank AML and anti-fraud programs 

AI isn’t a buzzword anymore; it’s a core operational layer. Banks globally see the value, with 89% actively encouraging its adoption. And the momentum is building with 70% of institutions using AI to some extent in financial crime and compliance. If we look at where FCC teams sit on the adoption curve, nearly half are piloting, 16% are operationally deployed, and 6% have embedded AI at strategic scale.  

Right now, fraud prevention leads with 10% having deployed it at scale, another 23% operational and 32% piloting. AML transaction monitoring is a close second. 5% are using AI at scale, 17% are operational and 36% are piloting.  

The data shows AI adoption in FCC will only grow. Investment follows conviction with 82% of banks planning to grow AI spending by more than 25% in the next year or two. 

The key takeaway: If you're not at least piloting AI, you're falling behind. 

 

AI saves more money than you think  

Banks that invested in AI for AML have found it saved more than expected. Only 26% thought cost reduction would be a top AI benefit, but 71% have already seen cost savings. These are not just marginal gains, 48% have saved over $1 million in the past year alone, and more than half expect annual savings to exceed $5 million in 2026. 

The key takeaway: As institutions scale their AI use, cost savings will grow. We’re seeing a drive to scale AI use firsthand: 92% of banks plan to boost their AI investment in financial crime and compliance over the next 2–3 years. 

 

Agentic AI isn’t just hype – it’s the next transformative technology

 

Key considerations for banks adopting agentic AI 

Banking leaders are optimistic about agentic AI, but they’re also clear about the challenges that need to be considered. Their top concern (60%) is meeting regulatory and audit requirements. While most FCC teams don’t expect major changes to headcount, many do worry about losing human expertise. In fact, 51% rank over-reliance on automation as their second-largest concern. A further 49% say increased operational complexity and cost is their third concern, suggesting that many expect agentic AI to come with a significant price tag and added implementation hurdles. 

The key takeaway: The biggest challenge is whether bank FinCrime teams have the expertise and governance structure to deploy generative and agentic AI safely. Banks that prioritize explainable AI models can implement generative and agentic AI with confidence, both shortening time-to-value and maximizing the transformative impact. 

 

Benefits ahead for banks adopting agentic AI 

Ninety-four percent of banking leaders are confident AI agents will streamline investigations. This reflects the fact that investigations are resource-intensive, filled with manual work and data collection that intelligent systems are able to pre-empt or even automate, if desired. Another 90% ranked SAR narrative drafting in their top 5, which ties directly into that same investigative workflow. Beyond that, 77% expect faster research and context gathering, while 69% anticipate better detection optimization and response to signals. 

The key takeaway: Banking leaders see clear, immediate use cases for AI to accelerate the time-consuming investigative work that currently drains resources.  

Now, let's look at payments and fintechs. 

 

What FCC leaders at payment firms & fintechs need to know 

 

Fraud prevention leads AI adoption  

Fraud is a massive cost center for payment firms and fintechs (chargeback fees, transactional losses etc)—one that AI can have a big impact on. Seventy three percent of firms are either operational or at the pilot stage of using AI for fraud prevention. This outpaces AML (71%) and screening (57%), reflecting where business pains are most acute. Natural Language Processing (NLP) drives much of this progress with 57% using NLP to capture more fraudulent transactions. 

The key takeaway: Successful payment firms and fintechs start with AI use cases that have the most urgent business impact. Trying to deploy AI everywhere at once simply dilutes ROI. 

 

Firms are saving with AI, but there's bigger returns ahead 

Payment firms are already reaping financial benefits: 73% report cost savings from AI in AML operations, with nearly a third saving $1m–$4.9m. But the biggest gains lie ahead: 42% expect savings exceeding $5m over the next 12 months, with 30% projecting $5m–$9.9m annually. Firms are putting their budgets behind that confidence with 74% planning to increase their investment, particularly in GenAI (88%) and agentic AI (84%). 

The key takeaway: Payment firms have already reduced operational costs with AI, and they expect the savings to grow as they scale up investment. These savings, from the application of AI in financial crime and compliance-related use cases, can unlock growth, allowing firms to scale more with existing resources. 

 

AI benefits are surpassing payment and fintech firms’ expectations  

AI promised payment firms improved detection accuracy, and it delivered. Seventy-four percent anticipated it as a top benefit, and 62% confirmed it as their biggest benefit realized. Payment firms process massive transaction volumes at thin margins, so reducing false alerts means fewer blocked legitimate transactions, less customer friction, and lower operational costs. The close second was faster alert triage at 59% meaning firms can quickly distinguish genuine fraud from legitimate activity. 

The key takeaway: AI is giving payment firms exactly what they’ve needed: stronger compliance and real commercial value.  

 

Get more stats and insights in the report  

This research shows that banking, payments, and fintech leaders have reached a turning point with AI. They’re already seeing impact. The challenge now is scaling it by using different AI technologies in a more unified way.  

The next phase is moving from experimentation to structured rollout. That means building internal expertise, integrating AI more deeply into existing systems, and keeping models explainable, well-governed, and cost-efficient as they go live. 

If you haven't yet explored our full reports, now's the time: 

  • Banking report: Discover how banks are using AI to transform the speed, precision, and scalability of their financial crime & compliance programs.
  • Payment & fintech report: Learn how payment and fintech's are using AI to improve fraud detection, streamline investigations, and reinforce compliance. 
     

The Future of FCC: AI Agents 

Thinking about how agentic AI could work for your institution? We wrote a practical guide exclusively for FCC leaders. It details how agentic AI differs from traditional and generative AI, and elaborates on specific use cases for agentic AI, covering:   

  • Improving investigations
  • Enhancing system accuracy
  • Optimizing workflows 

 Download it here. 

 

Share this page